ezyang

10 ms·

1.
by ezyang 2mo ago
One of the reasons why I have been thinking and writing this topic is because there is indeed a malaise among the team, which is along the lines of "AI agents are blowing up coding everywhere, what place does a compiler have in this br
2.
by ezyang 8mo ago
I just want to point out that Opus 4.5 actually knows this trick and will write the code to decode the IDs if it is working with GitHub's API lol
3.
The Gap Between a Helpful Assistant and a Senior Engineer(blog.ezyang.com)
2 points by ezyang 9mo ago|1 comments
4.
by ezyang 1y ago
The happy path way of getting code out of Codex is a PR. This is emphatically not true for Cursor.
5.
by ezyang 1y ago
Lmarena isn't that useful anymore lol
6.
by ezyang 1y ago
I quite liked this article, actually, and I'm quite an MCP stan. These all seem like legitimate problems that the burgeoning ecosystem needs to figure out. Some of them will be logically solved inside the MCP spec, but I also think som
7.
by ezyang 1y ago
I mean, obviously you want to run the local CI in some isolated way. But it's not such a bad idea for many projects. At PyTorch, we have the reverse problem, it's basically infeasible to run the CI locally, you really do need the
8.
Maintain a personal LLM coding benchmark(blog.ezyang.com)
1 points by ezyang 1y ago|0 comments
9.
by ezyang 1y ago
No, in fact, codemcp can be thought of as a fancier version of the official filesystem MCP that Anthropic released. It's 100% MCP.
10.
by ezyang 1y ago
One thing I'll say, is that if I was going to make people pay API costs (like cline/claude code) I probably wouldn't actually make an MCP. The MCP box is pretty limiting, and I'm only willing to pay the cost because tha
11.
by ezyang 1y ago
I definitely agree that for current models, the problem is finding where the LLM has comparative advantage. Usually it's something like (1) something boring, (2) something where you don't have any of the low level syntax or domai
12.
by ezyang 1y ago
One problem with keeping the changes separate is the LLM usually wants to test the code with the incremental new changes. So you need a working tree that has all the new changes. But then... why not use the real one?
13.
by ezyang 1y ago
So I have liked the Claude Pro style rate limiting over credits that refresh monthly but mostly because I only get snatches of 1-2 hours while I am on baby leave so I never actually get rate limited. As for learnings, I put a lot of them in
14.
by ezyang 1y ago
The thing is that, as many junior engineers can attest, randomly blundering around can still give you something useful! So you definitely can get value out of AI coding with the current generation of models.
15.
by ezyang 1y ago
I have used Cursor and my own MCP codemcp. Cursor has a lot of nice QoL that you can't get from an MCP package; the TAB is really good for traditional coding. Haven't used copilot so I don't have a comparison there. Definitel
16.
by ezyang 1y ago
When I posted this to Reddit there was a pretty lively discussion there: https://www.reddit.com/r/ChatGPTCoding/s/wRmnREUWzn
17.
by ezyang 1y ago
Cline has famously huge API token costs as it is profligate with context. Because codemcp plugs into Claude Desktop you only pay for your Claude Pro sub, similar to Cursor's pricing model
18.
by ezyang 1y ago
In my experience, the model is king, and codemcp will operate fairly similarly to Cursor with the same failure modes as Cursor due to Sonnet 3.7. One thing that I like about codemcp is I can customize aspects of the interaction as I discove
19.
by ezyang 1y ago
Yeah, Sonnet 3.5/3.7 are doing heavy lifting. Maybe the SOTA Gemini models would do better, I haven't tried them. Generating correct patches is a funny minigame that isn't really solved, despite how easy it is to RL on.
20.
by ezyang 1y ago
In fact, the test framework I was using at the time (jest) did in fact support this. But the person who had originally written the tests hadn't had the foresight to use snapshot tests for this failing test!
21.
by ezyang 1y ago
They are listed on one page right now! Haha
22.
by ezyang 1y ago
I'm surprised you managed to get Desktop running on Linux lol. You don't need the filesystem/git MCPs alongside codemcp; in fact, it's better not to have them so that Claude consistently uses' codemcp's equiva
23.
by ezyang 1y ago
Hi Hacker News! One of the things about this blog that has gotten a bit unwieldy as I've added more entries is that it's a sort of undifferentiated pile of posts. I want some sort of organization system but I haven't found on
24.
by ezyang 1y ago
And if you don't like clicking, even once, check out my other project, Refined Claude :)
25.
by ezyang 1y ago
Most of the problem is installing the MCP servers, which is more annoying on Windows. https://github.com/ezyang/codemcp#getting-started has instructions that I've personally tested for installation on Windows, whi
26.
by ezyang 1y ago
I don't really recommend using filesystem MCP directly, it won't checkpoint changes so it's easy to end up in a state where you can't recover an older working version of the code. Use an actual coding oriented MCP.
27.
by ezyang 1y ago
But only Claude Desktop gets flat $20 pricing from Claude Pro lol
28.
by ezyang 1y ago
Not built in, you'll have to use something like https://github.com/sparfenyuk/mcp-proxy
29.
by ezyang 1y ago
And with Claude Code, it's typically not $0.08... it's more like $0.50, $5.00 just for a roll LOL. Variable rewards gambling addiction? Definitely...
30.
by ezyang 1y ago
If you'd done it with MCP it would only have cost you $20 and you would still have had the rest of the month to use your Claude Pro sub :P

More ›